A new approach for deconvolution and filtering of 3-D microscopy images

A new approach to the deconvolution and filtering of 3-D microscopy images is introduced in this paper. A state-space representation of the image is derived according to the assumption that the whole image can be modelled by an ensemble of smooth 3-D Gaussian random fields. Blurring and noise are then easily included in the representation. Making use of this model the image restoration is carried out by means of a Kalman-based minimum variance estimation algorithm. The reported simulation results show high performances of the proposed approach.